Piping programs cannot manage every component individually, so they group. A circuit is the unit of that grouping: a run of piping treated as one item for corrosion evaluation, CML statistics, and interval setting. The convenience is enormous, and so is the hidden claim. Grouping components into a circuit asserts that they corrode together: same mechanisms, similar rates, one population whose statistics mean something. Circuitization is defensible exactly to the extent that assertion is true and documented.
What makes a valid circuit
The grouping logic is corrosion logic, not drawing logic. Components belong in one circuit when they share:
- Process service and conditions. Same fluid, and temperature and phase behavior similar enough that the corrosion environment is genuinely common. A line that crosses a dew point or a major temperature gradient is crossing a circuit boundary, whatever the line number says.
- Material of construction. Different metallurgy responds differently to the same service. Carbon steel and stainless segments of the same line are different circuits.
- Credible damage mechanisms. The mechanism review, with API 571 as the reference, should read the same for every component in the group. One mechanism list per circuit is the whole point.
- Expected corrosion behavior. Similar enough rates that pooling the CML data produces a meaningful distribution rather than a blend of two populations.
Line lists and P&ID boundaries are where circuitization usually starts, and they are the wrong stopping point. Drawings encode construction logic. Circuits encode corrosion logic. The two align often enough to be tempting and diverge often enough to matter.
The boundaries that break programs
Certain features force circuit boundaries regardless of how the drawings read, because they change the corrosion environment mid-line:
- Injection and mixing points, which create localized regimes aggressive enough that the codes treat them as their own inspection problem with their own coverage.
- Phase changes, condensation fronts and flashing points, where the wetted chemistry differs from everything upstream.
- Dead legs and intermittent-flow segments, which corrode on stagnation logic while the flowing main corrodes on velocity logic.
- Material and specification breaks, including changes in corrosion allowance.
Lumping these into a parent circuit does statistical damage in both directions. The aggressive segment's readings are diluted by the healthy majority, so its true rate is understated. Meanwhile the circuit's scatter inflates, so confidence in the pooled rate degrades. The result is a program that inspects the wrong places with misplaced confidence, wearing a tidy circuit list.
The record behind the grouping
Because circuitization is a set of engineering claims, the claims need files. For each circuit: the component extent tied to drawings, the service basis, the material, the mechanism list, and the reasoning at every boundary, especially the boundaries that cut across line numbers. Revisions matter as much as the original. When a process change or an inspection finding shows two halves of a circuit behaving differently, the defensible response is to split the circuit on the record, re-home the CML history to the new structure, and recompute rates per side.
That re-homing is the step that fails in spreadsheet programs. Splitting a circuit means every historical reading must land in the correct successor, or years of rate history become unusable at exactly the moment it matters most.
intelliSPEC keeps CML history anchored through circuit restructuring, so a split refines the data instead of orphaning it. Bring your worst circuit to a demo and re-draw it against live history.
